RE: URI aliases: suggested text for webarch 2.3.1

The problem is endemic [1] to the web architecture 
in that it conflates the concepts of name and address. 
The resolution to a sublocation in a resource should 
not and is not a problem of the web infrastructure; 
just the theories about it.

Dynamic (a URI that returns a changing state of a value 
such as a clock tick) and static (a URI that returns 
a state asserted to be persistent such as a zip code) 
states should not be confusing, but the conflation 
of names and addresses ensures it will be in the 
explanation if not in the implementation.

len

[1] http://www.thefreedictionary.com/endemic word sense 3

-----Original Message-----
From: www-tag-request@w3.org [mailto:www-tag-request@w3.org]On Behalf Of
Thompson, Bryan B.
Sent: Tuesday, August 10, 2004 2:38 PM
To: Roy T. Fielding; www-tag@w3.org
Subject: RE: URI aliases: suggested text for webarch 2.3.1



Roy,

Is this meant to describe a situation in which a complex "resource" is
revealed through a
set of URIs, all of which encode some state that lets the server
disambiguate which view
of the resource to send to the client.

For example, the entire telephone directory is available, perhaps as
RDF/XML, from:

http://www.mytelco.com/phonebook

However using one mechanism or another, the server is willing to send a
representation of
all phonebook information for a person, business, zip code, city, state,
etc.  E.g.,

http://www.mytelco.com/phonebook/people/Your%20Name
http://www.mytelco.com/phonebook/biz/Your%20Biz%20Name
http://www.mytelco.com/phonebook/zip/20009
http://www.mytelco.com/phonebook/state/CA

So, these URIs all "alias" the parts of the state (a phone book) on the
server.  Is this
something that the text, below, is meant to be critical of?

What I think is happening is that it is difficult to talk about the referent
of the URI
and therefore it is difficult to talk about URI aliases.  If two URIs reveal
precisely
the same state, then fine.  However I would consider the examples above
partially aliased
state in the sense that using PUT to update a person's information,
/people/foo, would
cause updates to be reflected in the entire phone book, in the view for the
zip code in
which they live, etc.

Thanks,

-bryan

-----Original Message-----
From: www-tag-request@w3.org [mailto:www-tag-request@w3.org] On Behalf Of
Roy T. Fielding
Sent: Tuesday, August 10, 2004 3:06 PM
To: www-tag@w3.org
Subject: URI aliases: suggested text for webarch 2.3.1



[URI aliases, 2.3.1]

URI aliases are harmful when they cause bifurcation in the web of related
resources.  A corollary of Metcalfe's Principle (the "Network Effect") is
that the value of a given resource can be measured by the number and value
of other resources that link to it (the network neighborhood of the measured
resource). This type of valuation is commonly used to rank the relative
value of search results (e.g., Google) because people tend to create links
relating a given topic to those resources that they feel best reflect that
topic, and hence the number of inbound references are a reflection of the
degree to which the community values a resource.  The problem with aliases
is that if half of the neighborhood points to one URI for a given resource,
and the other half points to a second, different URI for that same resource,
the neighborhood graph splits (bifurcates). The aliased resource is not the
only one undervalued because of this split: the entire neighborhood of
resources become undervalued due to the missing second-order relationships
that should have existed among the referring resources by virtue of their
references to the aliased resource.

[note, aliases are also in 2.2 for some reason unknown]

....Roy

Received on Tuesday, 10 August 2004 19:48:12 UTC